Materials Research Bulletin 48 (2013) 1845–1850

Photocatalytic degradation of methyl orange dye in water solutions in thepresence of MWCNT/TiO2 composites

S. Da Dalt *, A.K. Alves, C.P. Bergmann

Department of Material, Federal University of Rio Grande do Sul, Av. Osvaldo Aranha 99, Laboratory 705C, ZIP 90035-190, Porto Alegre, RS, Brazil

A R T I C L E I N F O

Article history:

Received 12 July 2012

Received in revised form 2 January 2013

Accepted 7 January 2013

Available online 24 January 2013

Keywords:

A. Nanostructures

B. Sol-gel chemistry

C. Electron microscopy

D. Catalytic properties

A B S T R A C T

The textile and dyestuff industries are the primary sources of the release of synthetic dyes into the

environment and usually there are major pollutants in dye wastewaters. Because of their toxicity and

slow degradation, these dyes are categorized as environmentally hazardous materials. In this context,

carbon nanotubes/TiO2 (CNTs/TiO2) composites were prepared using multi-walled CNTs (MWCNTs),

titanium (IV) propoxide and commercial TiO2 (P251) as titanium oxide sources, to degrade the methyl

orange dye in solution through photocatalyst activity using UV irradiation. The composites were

prepared by solution processing followed by thermal treatment at 400, 500 and 600 8C. The

heterojunction between nanotubes and TiO2 was confirmed by XRD, specific surface area. The coating

morphology was observed with SEM and TEM.

� 2013 Elsevier Ltd. All rights reserved.

Contents lists available at SciVerse ScienceDirect

Materials Research Bulletin

jo u rn al h om ep age: ww w.els evier .c o m/lo c ate /mat res b u

1. Introduction

Dyes are one of the most hazardous chemical compound classesfound in industrial effluents and need to be treated since theirpresence in water bodies reduces light penetration, precluding thephotosynthesis of aqueous flora [1,2]. They are also estheticallyobjectionable for drinking and other purposes [3]. Dyes can causeallergy, dermatitis, skin irritation [4] and also provoke cancer [5]and mutation in humans [6].

Two efficient methods for the degradation of synthetic dyes arethe adsorption procedure [7] and the photocatalytic process [8,9].The first process transfers the dyes from the water effluent to asolid phase, thereby keeping the effluent volume to a minimum[10–12]. Subsequently, the adsorbent can be regenerated or storedin a dry place without direct contact with the environment [13]. Inthe other process, dye molecule degrades when irradiated byenergy.

Multi-walled carbon nanotubes (MWCNTs) can be consideredgood supports for materials with photocatalytic properties due totheir good mechanical properties, chemical stability and thepresence of mesopores, which favor the diffusion of reagentspecies [8]. The use of composites of MWCNTs and TiO2 cantherefore have very interesting potential applications in photo-catalysis.

In order to optimize the catalytic properties of TiO2, MWCNTshave been used as support, providing easier passage in the electron

* Corresponding author.

E-mail address: silvana.da.dalt@ufrgs.br (S. Da Dalt).

0025-5408/$ – see front matter � 2013 Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.materresbull.2013.01.022

transfer process and, therefore, increasing the catalytic activity oftitania [14]. Among the oxide semiconductor photocatalysts,titania (TiO2) is a great option due to its high oxidizing properties,absence of toxicity and photostability. It is believed that thedispersion of TiO2 on the surface of MWCNTs favors theappearance of many active sites for photocatalytic degradation[15].

According to Leary and collaborators [14], the performance ofMWCNTs in the presence of TiO2 in photocatalysis, can beunderstood as the CNT acting as a receptor or electron donorunder UV irradiation, forming a very reactive superoxide orhydroxyl radical. These ions react on the surface of the oxide andbecome responsible for the degradation of organic compounds.

The recombination of the electron–hole pair formed in thegeneration of superoxide radical and hydroxyl ions is undesirablein photocatalytic reactions. The addition of electron acceptors caninhibit the recombination of the pair since it increases the amountof electrons confined to the conduction band, avoiding therecombination and generating more radicals and oxidants. It isbelieved that CNTs can hinder the recombination processes.

Photocatalytic applications of wide reaching importanceinclude water splitting for hydrogen generation, degradation ofenvironmental pollutants in aqueous contamination and waste-water treatment, carbon dioxide remediation, self-cleaningactivity and air purification [14,16].

In the present study, MWCNTs/TiO2 composites were obtainedand used to degrade organic dyes in water. The methyl orange(MeO) dye was used in aqueous media as a model dye to evaluatethe photocatalytic activity of the CNTs/TiO2 composites. Theadsorption capacities in the dark and the photocatalytic activities

Fig. 1. XRD pattern of the MWCNT/TiO2 composites, P25 used as reference for the

presence of pure TiO2 and MWCNT.

S. Da Dalt et al. / Materials Research Bulletin 48 (2013) 1845–18501846

of the CNTs/TiO2 composites under UV light were studied as well asthe structural characterization of the composites.

2. Experimental design

2.1. Synthesis

The following precursors were employed for the functionaliza-tion process of MWCNT with TiO2: titanium (IV) propoxideTi(OPr)4 (TPP), as alkoxide precursor of TiO2, provided by Sigma–Aldrich; commercially available TiO2 (AEROXIDE1 – P25) made byEvonik; nitric acid P.A., produced by Synth; isopropyl alcohol anddeionized water. The Baytubes1 were supplied by Bayer1, andwere employed as the MWCNT source.

The composites were prepared by a modified sol–gel method.The production of the samples was basically divided into twosystems: (i) using TPP and (ii) using P25. The molar ratio forsystem (i) and (ii) is 6:140:5:1 = Ti(OPr)4/P25:R-OH:H2O:HNO3,with 0.33 g of MWCNT being mixed in 29 ml of solution for eachsample. The liquid systems containing precursors were keptunder magnetic stirring and heated up to 40 8C for 1 h. After thisperiod, MWCNTs were added to the liquid systems, whichremained under stirring for another hour. The process wascompleted before of the samples started the gelation process, i.e.there was no complete transformation of sol in gel. Thereafter, theliquid systems were filtered and the material was kept at 100 8Cfor 24 h. The samples of system (i), called T4, T5 and T6 weresubjected to heat treatment at 400 8C, 500 8C and 600 8C for 1 h,respectively, using a heating rate of 2.5 8C in a muffle type furnacein air atmosphere. The heat treatment of the sample of system (ii),called PA, was not necessary because the precursor was alreadycrystallized.

2.2. Characterization

Scanning electron microscopy (SEM – JEOL JSM-6060) operat-ing between 12 and 15 kV, and transmission electronic microscopy(TEM – JEM 1200EXII-120 kV) operating to 80 kV were employedfor the morphological analysis of the samples. The crystallinity ofthe composites obtained was evaluated by the X-ray diffraction(XRD) technique using a PHILIPS diffractometer (Model X’PertMPD) equipped with a graphite monochromator and a copperanode, operating at 40 kV and 40 mA. Analyses were performed ina 2u range of 20–708, with steps of 0.058 for 2 s, with Cu Karadiation. The specific surface area was determined by theBrunauer–Emmett–Teller (BET) method, using N2 as gas adsorp-tion media, using a Quantachrome Autosorb Automated GasSorption System (NOVA 1000). Thermogravimetric analysis (TGA)of the NTCPMs–TiO2 composites were performed in a Mettlerapparatus (A851e) using an O2 atmosphere and with a heating rateof 10 8C/min until the temperature of 900 8C, with alumina as thereference material.

2.3. Photocatalytic activity

The photocatalytic activity of the composites was evaluated byfollowing the decomposition of the methyl orange dye (MeO)under UV light. Initially, the MWCNTs/TiO2 composites were addedto a MeO solution and kept in the dark for 1 h. Subsequently, thesolution was dispersed using ultrasound for 10 min. After thisperiod, the solution was irradiated with UV light. The photo-catalytic activity of MWCNT/TiO2 composites, as well as of theMWCNT/P25 samples, was evaluated using 12 UVA lamps(wavelength 365 nm or 3.39 eV) of 8 W each. The dye concentra-tion was set at 1.0 � 10�5 mol/L. The concentration of compositesin suspension was 0.4 g/L and 0.8 g/L in the MeO solution. The first

dispersed dye sample was collected at the end of the adsorptionperiod in the dark, before the sample was exposed to UVirradiation.

The reaction recipient consisted of a quartz Dreschel bottlefitted with a silicone septum to facilitate withdrawal of samplesfrom the reaction dispersion. Samples were collected periodicallywith a syringe from the reactor every 10 min during a 70 min timeinterval and then filtered (pore diameter 0.2 mm) into a 4 mLPMMA cuvette. The absorbance at the wavelength of 465 nm wasrecorded (Bioespectro SP 200) for each sample containing thecomposites and the MeO in a solution with pH 7.

3. Results and discussion

XRD spectra (Fig. 1) were collected to analyze the effect of CNTson the crystallization of the nanocomposites. A sample of pureTiO2-P25 shows the anatase phase with well-defined peaks andsome peaks of the poorly crystalline rutile-type phase. However, inthe presence of CNTs its characteristic crystallinity apparentlydecreases, as can be observed in the XRD spectra of the PA sample(Fig. 1). A similar effect occurs with the MWCNT samples: in thepresence of TiO2, apparently there is a decrease in the intensity ofthe CNT characteristic peak.

According to Wang and collaborators [17], the disappearance ofcharacteristic CNT peaks in the XRD patterns of the compositematerials is consistent with a homogeneous coverage of TiO2 onMWCNT, which is additionally supported by the absence ofMWCNT aggregated pores in the composite materials. On the otherhand, the introduction of MWCNT to TiO2 favors less extendedcrystallized TiO2 domains on the MWCNT surface, thus avoidingTiO2 particle agglomeration. All factors account for the increase insurface areas of the composite materials.

The 2u positions at 268 and 43.48, on the (0 0 2) and (1 0 1)planes, respectively, correspond to characteristic peaks of CNTs[8]. These peaks can be visualized on the spectra of the MWCNTsample, which has no TiO2. The main peak of the anatase phaseis observed at 25.48 on the (1 0 1) plane [15,18]. This couldexplain the overlap of peaks on this position in the T4, T5 and T6samples, which hinders the observation of the typical diffractionpeak of graphite at 268 corresponding to the (0 0 2) plane[19,20].

In general, the amorphous TiO2 shows lower photocatalyticactivity. However, if it is heat treated at 450 8C in air, part of thematerial can be transformed into the anatase phase and promotephotocatalytic activity [17].

Table 1Specific surface area (SBET) of the MWCNT, TiO2-P25 and TiO2 composites.

Sample Temperature of thermal treatment in air (8C) SBET (m2/g)

MWCNT – 225.49

T4 400 122.27

T5 500 194.00

T6 600 150.85

PA – 63.78

P25 – 68.07

Table 2Thermal stability of the samples according to TGA.

Sample Weight loss (%)a Sample color TiO2 remaining mass (g)

PA 55.25 0.023

T4 48.05 0.026

T5 43.08 0.029

T6 2.49 0.048

P25 1.50 0.049

MWCNT 90.30 –

a Final mass values obtained at 900 8C.

S. Da Dalt et al. / Materials Research Bulletin 48 (2013) 1845–1850 1847

Specific surface area (Table 1) is a significant surface parameterfor MWCNTs and MWCNTs/TiO2 nanocomposites. It is an impor-tant parameter for evaluating the surface characteristics ofMWCNTs/TiO2. It also influences the performance of photocatalyticactivity since this is associated with a high specific surface area.The increase of TiO2 on the surface of MWCNTs, causes a decreasein micropore volume and specific surface area. This observationcan be explained by the oxide layer that blocks the micropores ofMWCNTs [16].

MWCNTs show a specific surface area (SBET) of 225.49 m2/g,while the surface area of commercial TiO2-P25 was 68.07 m2/g. Thecomposite samples show higher SBET, when compared to the P25sample, except for the PA sample. The PA sample presents theanatase phase and according to the XRD this can be associated tothe decrease in specific surface area and crystallinity. MWCNTs/TiO2 prepared from the TPP precursor show a specific surface arearanging from 122.27 to 194 m2/g. The presence of MWCNTs helpsto significantly increase the specific surface area of the compositessamples, which leads to higher adsorptive capabilities [17].

Fig. 2 shows the thermogravimetric analysis (TGA) of samplesP25 MWCNTs and MWCNTs–TiO2. The composites obtained fromTPP were analyzed after heat treatment. TGA was used to evaluatetemperature and oxidation weight loss conditions of pureMWCNTs and those combined with TiO2. According to Fig. 2,the pure MWCNTs have significant mass loss between 480 8C and570 8C, these temperatures corroborate the oxidation values forMWCNTs described in the literature [21]. However, one can seethat this temperature is increased when the nanotubes areassociated with TiO2.

The T4, T5 and PA composites show weight loss at highertemperatures: 550 8C and 610 8C. For all samples, except theMWCNT sample, a weight loss of around 10% between 20 8C and500 8C is observed, possibly due to the loss of organic matter of theTiO2 precursor and of other organics from the functionalization

Fig. 2. Thermogravimetric analysis (TGA) of MWCNT/TiO2 composites, MWCNT and

P25.

reagent that were not completely eliminated during the heattreatment. After 630 8C, there was no significant loss of mass,which indicates that only TiO2 was present and that the nanotubesin the samples were decomposed. The thermal behavior allowed usto evaluate weight loss, as shown in Table 2. The amount ofremaining mass was assigned to the amount of TiO2 present in thesample.

Samples P25 and T6 experienced the smallest loss in mass dueto the low CNT loads. The larger mass present in these samples wastherefore attributed to TiO2. Apart from the P25 sample, the T6sample experienced the smallest weight loss among the compo-sites, i.e. 2.49%. The remaining mass, 97.5%, can be assigned to TiO2.The greatest amount of weight loss was observed in the PAcomposite, 55.25%, with a remaining mass of 44.75%, assigned toP25. The composites underwent greater mass loss due to theoxidation of organic material, especially those that containedlarger amounts of CNTs. This is especially true for the PA samplesince it was not heat treated and thus contained approximately thesame amount of nanotubes that were added to the liquid systemsduring the production process.

The SEM images of MWCNTs/TiO2 are shown in Fig. 3. The PAand T4 samples show similar morphologies. The image observedfor the T4 sample is in line with the large density of TiO2

agglomerations with cauliflowerlike shapes [22]. The T5 samplereveals a spherical agglomeration morphology, indicating thepreponderant presence of TiO2 in the sample, possibly due to thedecomposition shrinkage and densification phenomenon whichaccompanies the crystallization of TiO2.

Oh [23] and Chen and collaborators [16] stated that a higherphotocatalytic activity of MWCNTs/TiO2 composites can beattributed to the homogeneous distribution of CNTs and animportant role as energy sensitizer to improve the quantumefficiency and charge transfer of carbon nanotubes. The morphol-ogy and particle distribution of TiO2 on CNTs can be confirmed byTEM (Fig. 4). The TiO2 grain shape is clearly visible in Fig. 4(a),which is observed the appearance spherical the particles withaverage diameter small than 30 nm, due to coated by P25 wellcrystalline according XRD. On the PA sample, the TiO2 is physicallyaggregate on the CNT surface, because there was no heat treatmenton this sample. Hence a covalent bond between P25 and CNTshould not be expected. However in Fig. 4(b), corresponding to theT4 sample, it is constituted the larger agglomeration the TiO2

Fig. 3. SEM micrograph images of the (a) PA (b) T4 and (c) T5 samples.

Fig. 4. TEM images of the PA (a) and T4 (b) samples.

Fig. 5. Photocatalytic activity of pure MWCNT, pure TiO2-P25 and MWCNT/TiO2

composites (T4, T5 and T6), MWCNT–P25 (PA) using 0.05 g of catalyst in the

solution at pH 7.

S. Da Dalt et al. / Materials Research Bulletin 48 (2013) 1845–18501848

particles and non constituted by particles shapes defined, due tocoating from TPP and the treatment temperature inadequate forcomplete crystallization.

Figs. 5 and 6 show the results obtained for the evaluation of thephotocatalytic activity of different amounts of catalysts in thesolution: 0.05 and 0.1 g, respectively, of pure MWCNTs, pure TiO2-P25 and MWCNT/TiO2 composites (T4, T5 and T6) and MWCNTs–P25 (PA), after 1 h in the dark.

In Fig. 5 one can observe that the standard TiO2-P25 sample ismore photocatalyticly active than its CNT composite (PA sample).There is, however only 0.023 g of P25 present in the PA sample,while the total content of catalyst in the dye solution was 0.05 g,

according Table 2, or possibly an insufficient coverage of theMWCNT with TiO2 may be attributed [24]. The T5 sample is moreefficient than the T4 sample, possibly by temperature of heattreatment closest to complete formation of anatase phase and

Fig. 6. Photocatalytic activity of pure MWCNT, pure TiO2-P25 and MWCNT/TiO2

composites (T4, T5 and T6), MWCNT–P25 (PA) using 0.1 g of catalyst in the solution

at pH 7.

S. Da Dalt et al. / Materials Research Bulletin 48 (2013) 1845–1850 1849

because the TiO2 load of the T5 sample is slightly greater (0.029 g)than the TiO2 load of the T4 sample (0.026 g).

The T6 sample does not present active photocatalytic material,although this sample show higher loading of TiO2 from thecomposite (0.048 g), possibly associated with low mass ofnanotubes in this sample, since this sample had a weight losscarbonacea to 570 8C, and the low concentration of carbon wouldcause the scattering of light during the photocatalytic analysis.When considering Fig. 6, however, where the amount of catalyst inthe solution is increased, the activity observed in the PA sample ishigher than that of pure P25. This can be attributed to thecombined effect of several possible concomitant factors, such asthe formation of NTCs/TiO2 heterojunctions that could reduce therate of recombination of photoinduced electrons and holes, and theabsorption of photons from the surface of the nanotube. As a result,the injection of electrons in the conduction band of TiO2 leads tothe formation of free radicals (superoxide and hydroxyl radicals).The T5 sample shows a similar behavior to the PA sample,enhancing the similarity between the phase of TiO2 in T5 and PA.

The behavior of pure MWCNTs in Figs. 5 and 6 may beassociated with adsorption effects caused by their higher specificsurface area when compared to other samples, resulting, therefore,in a greater adsorption capacity for reactive species.

Studies [14] using suspensions of MWCNTs/TiO2 have sug-gested that the function of CNT in the composite photocatalyst canbe attributed to three distinct mechanisms: (i) CNTs can act as ameans of dispersion of TiO2 nanoparticles, (ii) CNTs can act as a co-adsorbent, and (iii) CNTs act as a photosensitizer. The firstmechanism is more significant when the TiO2 particles aregenerated simultaneously during the synthesis of the composite.In this case, the chemical groups on the surface of the CNTs may actas anchor points for the TiO2 nanoparticles [25].

When the samples made with TPP are compared, sample T5(Fig. 6), has the best catalytic performance in relation to othersamples. This result may be associated with the higher specificsurface area of this sample and an optimized TiO2 load on the CNTsurface, for this sample show 0.029 g of TiO2 according Table 2.However, the load of TiO2 on the T6 sample is not related to theabsence of photocatalytic activity. Sampaio and collaborators [25]observed that the introduction of CNTs leads to a decrease in theband gap energy. The increase of the heat treatment temperaturecould be associated with an increase of the band gap, thus notreaching the energy of radiation UV used in photocatalysis.

From the photocatalytic results, it is possible to infer from thevariation in the amount MWCNTs/TiO2 in the reactive solution that

there is different photocatalytic behavior among samples. Theaddition of 0.1 g of composite MWCNTs/TiO2 may favor theseparation of an electron–hole pair, increasing the photocatalyticactivity, but can also favoring the adsorption mechanism of highamounts of CNTs [26]. However, Zhang and collaborators [19]obtained CNT/TiO2 composites from titanium butoxide as TiO2

precursor using a modified sol–gel technique. The results showedthat a load higher CNT improved the methylene blue degradationin solution. Jiang and collaborators [18] observed that a higherphotocatalytic activity was obtained on pH 7 solution during aphotocatalysis. These results are in agreement with the resultsobtained in this study.

The interaction between CNTs and TiO2 composite materialsalso depends on certain aspects, such as the size of TiO2 particles,the crystalline phase composition of TiO2 and surface chemistry ofthe MWCNT/TiO2 compound [14].

When employing the photocatalyst in suspension, the optimalCNT load corresponds to the amount of CNT in the composite thatwould promote a more active catalyst. In other words, the optimalload is the one where a maximum of synergy between MWCNTsand TiO2 is achieved. Larger quantities of MWCNTs can causescattering of light, leading to a decrease in the photo-efficiency ofthe process [25].

4. Conclusion

The photoactivity of the prepared materials was evaluatedthrough the degradation of the methyl orange dye (MeO) in anaqueous solution under UV irradiation. The photocatalytic effect ofCNT/TiO2 nanocomposites occurs not only because of theadsorption of MWCNTs, but also because of the electron transferbetween MWCNTs and TiO2, removing the dye from the solution.The PA sample showed satisfactory results for photocatalyticactivity, which can be compared to the pure, commercial TiO2 (P25sample). The TiO2 load should be considered for each sample. Inthis case, the loading of TiO2 of the PA sample is approximately halfof that of the P25 sample. The photocatalytic activity, however, wassimilar. This shows the contribution of CNTs to photocatalyticperformance, even though other factors are involved in photo-catalytic activity, such as the conditions of synthesis, TiO2 particlesize, the nature of the CNTs, pore distribution and the compositionof phases.

References

[1] E.C. Lima, B. Royer, J.C.P. Vaghetti, N.M. Simon, B.M. Cunha, F.A. Pavan, E.V.Benvenutti, R.C. Veses, C. Airoldi, J. Hazard. Mater. 155 (2008) 536–550.

[2] B. Royer, N.F. Cardoso, E.C. Lima, J.C.P. Vaghetti, N.M. Simon, T. Calvete, R.C. Veses,J. Hazard. Mater. 164 (2009) 1213–1222.

[3] B. Royer, N.F. Cardoso, E.C. Lima, T.R. Macedo, C. Airoldi, J. Hazard. Mater. 181(2010) 366–374.

[4] D.S. Brookstein, Dermatol. Clin. 27 (2009) 309–322.[5] R.O.A. Lima, A.P. Bazo, D.M.F. Salvadori, C.M. Rech, D.P. Oliveira, G.A. Umbuzeiro,

Mutat. Res. Genet. Toxicol. Environ. Mutagen. 626 (2007) 53–60.[6] P.A. Carneiro, G.A. Umbuzeiro, D.P. Oliveira, M.V.B Zanoni, J. Hazard. Mater. 174

(2010) 694–699.[7] F.M. Machado, C.P. Bergmann, T.H.M. Fernandes, E.C. Lima, B. Royer, T. Calvete, S.B.

Fagan, J. Hazard. Mater. 192 (2011) 1122–1131.[8] Y. Yu, J.C. Yu, J.G. Yu, Y.C. Kwok, Y.K. Che, J.C. Zhao, L. Ding, W.K. Ge, P.K. Wong,

Appl. Catal. A: Gen. 289 (2005) 186–196.[9] J.M. Herrmann, C. Duchamp, M. Karkmaz, B.T. Hoai, H. Lachheb, E. Puzenat, C.

Guillard, J. Hazard. Mater. 146 (2007) 624–629.[10] T. Calvete, E.C. Lima, N.F. Cardoso, S.L.P. Dias, F.A. Pavan, J. Chem. Eng. 155 (2009)

627–636.[11] E.M. Cuerda-Correa, J.R. Domınguez-Vargas, F.J. Olivares-Marın, J.B. Heredia, J.

Hazard. Mater. 177 (2010) 1046–1053.[12] R. Malarvizhi, Y.S. Ho, Desalination 264 (2010) 97–101.[13] T. Calvete, E.C. Lima, N.F. Cardoso, S.L.P. Dias, E.S. Ribeiro, Clean Soil Air Water 38

(2010) 521–532.[14] R. Leary, A. Westwood, Carbon 49 (2011) 741–772.[15] A. Jitianu, T. Cacciaguerra, R. Benoit, S. Delpeux, F. Beguin, S. Bonnamy, Carbon 42

(2004) 1147–1151.

S. Da Dalt et al. / Materials Research Bulletin 48 (2013) 1845–18501850

[16] M.L. Chen, F.J. Zhang, W.C. Oh, New Carbon Mater. 24 (2) (2009) 159–166.[17] W. Wang, P. Serp, P. Kalck, J.L. Faria, Mater. Res. Bull. 43 (2008) 958–967.[18] G. Jiang, X. Zheng, Y. Wang, T. Li, X. Sun, Powder Technol. 207 (2011) 465–669.[19] K. Zhang, F.J. Zhang, M.L. Chen, W.C. Oh, Ultrason. Sonochem. 18 (2011)

765–772.[20] N. Bouazza, M. Ouzzine, M.A. Lillo-Rodenas, D. Eder, A. Linares-Solano., Appl.

Catal. B: Environ. 92 (2009) 377–383.[21] J. Cheng, et al. Mater. Chem. Phys. 95 (2006) 5–11.

[22] A.M.O. Zevallos-Marquez, M.J.S.P. Brasil, F. Likawa, A. Abbaspourrad, C. Verıssimo,S.A. Moshkalev, O.L. Alves, J. Appl. Phys. 108 (2010) 083501.

[23] W.C. Oh, M.L. Chen, Bull. Korean Chem. Soc. 29 (1) (2008) 159–164.[24] B. Gao, G.Z. Chen, G. Li Puma, Appl. Catal. B: Environ. 89 (2009) 503–509.[25] M.J. Sampaio, C.G. Silva, R.R.N. Marques, A.M.T. Silva, J.L. Faria, Catal. Today 161

(2011) 91–96.[26] W. Zhou, K. Pan, Y. Qu, F. Sun, C. Tian, Z. Ren, G. Tian, H. Fu, Chemosphere 81 (2010)

555–561.